ViewVC Help
View File | Revision Log | Show Annotations | Download File
/cvs/CBOR-XS/XS.pm
(Generate patch)

Comparing CBOR-XS/XS.pm (file contents):
Revision 1.3 by root, Sat Oct 26 11:08:34 2013 UTC vs.
Revision 1.13 by root, Tue Oct 29 15:56:31 2013 UTC

12 $perl_value = decode_cbor $binary_cbor_data; 12 $perl_value = decode_cbor $binary_cbor_data;
13 13
14 # OO-interface 14 # OO-interface
15 15
16 $coder = CBOR::XS->new; 16 $coder = CBOR::XS->new;
17 #TODO 17 $binary_cbor_data = $coder->encode ($perl_value);
18 $perl_value = $coder->decode ($binary_cbor_data);
19
20 # prefix decoding
21
22 my $many_cbor_strings = ...;
23 while (length $many_cbor_strings) {
24 my ($data, $length) = $cbor->decode_prefix ($many_cbor_strings);
25 # data was decoded
26 substr $many_cbor_strings, 0, $length, ""; # remove decoded cbor string
27 }
18 28
19=head1 DESCRIPTION 29=head1 DESCRIPTION
20 30
21WARNING! THIS IS A PRE-ALPHA RELEASE! IT WILL CRASH, CORRUPT YOUR DATA AND 31WARNING! This module is very new, and not very well tested (that's up to
22EAT YOUR CHILDREN! 32you to do). Furthermore, details of the implementation might change freely
33before version 1.0. And lastly, the object serialisation protocol depends
34on a pending IANA assignment, and until that assignment is official, this
35implementation is not interoperable with other implementations (even
36future versions of this module) until the assignment is done.
23 37
24This module converts Perl data structures to CBOR and vice versa. Its 38You are still invited to try out CBOR, and this module.
39
40This module converts Perl data structures to the Concise Binary Object
41Representation (CBOR) and vice versa. CBOR is a fast binary serialisation
42format that aims to use a superset of the JSON data model, i.e. when you
43can represent something in JSON, you should be able to represent it in
44CBOR.
45
46In short, CBOR is a faster and very compact binary alternative to JSON,
47with the added ability of supporting serialisation of Perl objects. (JSON
48often compresses better than CBOR though, so if you plan to compress the
49data later you might want to compare both formats first).
50
25primary goal is to be I<correct> and its secondary goal is to be 51The primary goal of this module is to be I<correct> and the secondary goal
26I<fast>. To reach the latter goal it was written in C. 52is to be I<fast>. To reach the latter goal it was written in C.
27 53
28See MAPPING, below, on how CBOR::XS maps perl values to CBOR values and 54See MAPPING, below, on how CBOR::XS maps perl values to CBOR values and
29vice versa. 55vice versa.
30 56
31=cut 57=cut
32 58
33package CBOR::XS; 59package CBOR::XS;
34 60
35use common::sense; 61use common::sense;
36 62
37our $VERSION = 0.02; 63our $VERSION = 0.06;
38our @ISA = qw(Exporter); 64our @ISA = qw(Exporter);
39 65
40our @EXPORT = qw(encode_cbor decode_cbor); 66our @EXPORT = qw(encode_cbor decode_cbor);
41 67
42use Exporter; 68use Exporter;
43use XSLoader; 69use XSLoader;
70
71use Types::Serialiser;
44 72
45our $MAGIC = "\xd9\xd9\xf7"; 73our $MAGIC = "\xd9\xd9\xf7";
46 74
47=head1 FUNCTIONAL INTERFACE 75=head1 FUNCTIONAL INTERFACE
48 76
163 191
164=head2 CBOR -> PERL 192=head2 CBOR -> PERL
165 193
166=over 4 194=over 4
167 195
168=item True, False 196=item integers
169 197
170These CBOR values become C<CBOR::XS::true> and C<CBOR::XS::false>, 198CBOR integers become (numeric) perl scalars. On perls without 64 bit
199support, 64 bit integers will be truncated or otherwise corrupted.
200
201=item byte strings
202
203Byte strings will become octet strings in Perl (the byte values 0..255
204will simply become characters of the same value in Perl).
205
206=item UTF-8 strings
207
208UTF-8 strings in CBOR will be decoded, i.e. the UTF-8 octets will be
209decoded into proper Unicode code points. At the moment, the validity of
210the UTF-8 octets will not be validated - corrupt input will result in
211corrupted Perl strings.
212
213=item arrays, maps
214
215CBOR arrays and CBOR maps will be converted into references to a Perl
216array or hash, respectively. The keys of the map will be stringified
217during this process.
218
219=item null
220
221CBOR null becomes C<undef> in Perl.
222
223=item true, false, undefined
224
225These CBOR values become C<Types:Serialiser::true>,
226C<Types:Serialiser::false> and C<Types::Serialiser::error>,
171respectively. They are overloaded to act almost exactly like the numbers 227respectively. They are overloaded to act almost exactly like the numbers
172C<1> and C<0>. You can check whether a scalar is a CBOR boolean by using 228C<1> and C<0> (for true and false) or to throw an exception on access (for
173the C<CBOR::XS::is_bool> function. 229error). See the L<Types::Serialiser> manpage for details.
174 230
175=item Null, Undefined 231=item CBOR tag 256 (perl object)
176 232
177CBOR Null and Undefined values becomes C<undef> in Perl (in the future, 233The tag value C<256> (TODO: pending iana registration) will be used
178Undefined may raise an exception). 234to deserialise a Perl object serialised with C<FREEZE>. See L<OBJECT
235SERIALISATION>, below, for details.
236
237=item CBOR tag 55799 (magic header)
238
239The tag 55799 is ignored (this tag implements the magic header).
240
241=item other CBOR tags
242
243Tagged items consists of a numeric tag and another CBOR value. Tags not
244handled internally are currently converted into a L<CBOR::XS::Tagged>
245object, which is simply a blessed array reference consisting of the
246numeric tag value followed by the (decoded) CBOR value.
247
248In the future, support for user-supplied conversions might get added.
249
250=item anything else
251
252Anything else (e.g. unsupported simple values) will raise a decoding
253error.
179 254
180=back 255=back
181 256
182 257
183=head2 PERL -> CBOR 258=head2 PERL -> CBOR
188 263
189=over 4 264=over 4
190 265
191=item hash references 266=item hash references
192 267
193Perl hash references become CBOR maps. As there is no inherent ordering 268Perl hash references become CBOR maps. As there is no inherent ordering in
194in hash keys (or CBOR maps), they will usually be encoded in a 269hash keys (or CBOR maps), they will usually be encoded in a pseudo-random
195pseudo-random order. 270order.
271
272Currently, tied hashes will use the indefinite-length format, while normal
273hashes will use the fixed-length format.
196 274
197=item array references 275=item array references
198 276
199Perl array references become CBOR arrays. 277Perl array references become fixed-length CBOR arrays.
200 278
201=item other references 279=item other references
202 280
203Other unblessed references are generally not allowed and will cause an 281Other unblessed references are generally not allowed and will cause an
204exception to be thrown, except for references to the integers C<0> and 282exception to be thrown, except for references to the integers C<0> and
205C<1>, which get turned into C<False> and C<True> in CBOR. 283C<1>, which get turned into false and true in CBOR.
206 284
207=item CBOR::XS::true, CBOR::XS::false 285=item CBOR::XS::Tagged objects
208 286
287Objects of this type must be arrays consisting of a single C<[tag, value]>
288pair. The (numerical) tag will be encoded as a CBOR tag, the value will
289be encoded as appropriate for the value. You cna use C<CBOR::XS::tag> to
290create such objects.
291
292=item Types::Serialiser::true, Types::Serialiser::false, Types::Serialiser::error
293
209These special values become CBOR True and CBOR False values, 294These special values become CBOR true, CBOR false and CBOR undefined
210respectively. You can also use C<\1> and C<\0> directly if you want. 295values, respectively. You can also use C<\1>, C<\0> and C<\undef> directly
296if you want.
211 297
212=item blessed objects 298=item other blessed objects
213 299
214Blessed objects are not directly representable in CBOR. TODO 300Other blessed objects are serialised via C<TO_CBOR> or C<FREEZE>. See
215See the 301L<OBJECT SERIALISATION>, below, for details.
216C<allow_blessed> and C<convert_blessed> methods on various options on
217how to deal with this: basically, you can choose between throwing an
218exception, encoding the reference as if it weren't blessed, or provide
219your own serialiser method.
220 302
221=item simple scalars 303=item simple scalars
222 304
223TODO 305TODO
224Simple Perl scalars (any scalar that is not a reference) are the most 306Simple Perl scalars (any scalar that is not a reference) are the most
225difficult objects to encode: CBOR::XS will encode undefined scalars as 307difficult objects to encode: CBOR::XS will encode undefined scalars as
226CBOR C<Null> values, scalars that have last been used in a string context 308CBOR null values, scalars that have last been used in a string context
227before encoding as CBOR strings, and anything else as number value: 309before encoding as CBOR strings, and anything else as number value:
228 310
229 # dump as number 311 # dump as number
230 encode_cbor [2] # yields [2] 312 encode_cbor [2] # yields [2]
231 encode_cbor [-3.0e17] # yields [-3e+17] 313 encode_cbor [-3.0e17] # yields [-3e+17]
253 335
254You can not currently force the type in other, less obscure, ways. Tell me 336You can not currently force the type in other, less obscure, ways. Tell me
255if you need this capability (but don't forget to explain why it's needed 337if you need this capability (but don't forget to explain why it's needed
256:). 338:).
257 339
258Note that numerical precision has the same meaning as under Perl (so 340Perl values that seem to be integers generally use the shortest possible
259binary to decimal conversion follows the same rules as in Perl, which 341representation. Floating-point values will use either the IEEE single
260can differ to other languages). Also, your perl interpreter might expose 342format if possible without loss of precision, otherwise the IEEE double
261extensions to the floating point numbers of your platform, such as 343format will be used. Perls that use formats other than IEEE double to
262infinities or NaN's - these cannot be represented in CBOR, and it is an 344represent numerical values are supported, but might suffer loss of
263error to pass those in. 345precision.
264 346
265=back 347=back
266 348
349=head2 OBJECT SERIALISATION
267 350
351This module knows two way to serialise a Perl object: The CBOR-specific
352way, and the generic way.
353
354Whenever the encoder encounters a Perl object that it cnanot serialise
355directly (most of them), it will first look up the C<TO_CBOR> method on
356it.
357
358If it has a C<TO_CBOR> method, it will call it with the object as only
359argument, and expects exactly one return value, which it will then
360substitute and encode it in the place of the object.
361
362Otherwise, it will look up the C<FREEZE> method. If it exists, it will
363call it with the object as first argument, and the constant string C<CBOR>
364as the second argument, to distinguish it from other serialisers.
365
366The C<FREEZE> method can return any number of values (i.e. zero or
367more). These will be encoded as CBOR perl object, together with the
368classname.
369
370If an object supports neither C<TO_CBOR> nor C<FREEZE>, encoding will fail
371with an error.
372
373Objects encoded via C<TO_CBOR> cannot be automatically decoded, but
374objects encoded via C<FREEZE> can be decoded using the following protocol:
375
376When an encoded CBOR perl object is encountered by the decoder, it will
377look up the C<THAW> method, by using the stored classname, and will fail
378if the method cannot be found.
379
380After the lookup it will call the C<THAW> method with the stored classname
381as first argument, the constant string C<CBOR> as second argument, and all
382values returned by C<FREEZE> as remaining arguments.
383
384=head4 EXAMPLES
385
386Here is an example C<TO_CBOR> method:
387
388 sub My::Object::TO_CBOR {
389 my ($obj) = @_;
390
391 ["this is a serialised My::Object object", $obj->{id}]
392 }
393
394When a C<My::Object> is encoded to CBOR, it will instead encode a simple
395array with two members: a string, and the "object id". Decoding this CBOR
396string will yield a normal perl array reference in place of the object.
397
398A more useful and practical example would be a serialisation method for
399the URI module. CBOR has a custom tag value for URIs, namely 32:
400
401 sub URI::TO_CBOR {
402 my ($self) = @_;
403 my $uri = "$self"; # stringify uri
404 utf8::upgrade $uri; # make sure it will be encoded as UTF-8 string
405 CBOR::XS::tagged 32, "$_[0]"
406 }
407
408This will encode URIs as a UTF-8 string with tag 32, which indicates an
409URI.
410
411Decoding such an URI will not (currently) give you an URI object, but
412instead a CBOR::XS::Tagged object with tag number 32 and the string -
413exactly what was returned by C<TO_CBOR>.
414
415To serialise an object so it can automatically be deserialised, you need
416to use C<FREEZE> and C<THAW>. To take the URI module as example, this
417would be a possible implementation:
418
419 sub URI::FREEZE {
420 my ($self, $serialiser) = @_;
421 "$self" # encode url string
422 }
423
424 sub URI::THAW {
425 my ($class, $serialiser, $uri) = @_;
426
427 $class->new ($uri)
428 }
429
430Unlike C<TO_CBOR>, multiple values can be returned by C<FREEZE>. For
431example, a C<FREEZE> method that returns "type", "id" and "variant" values
432would cause an invocation of C<THAW> with 5 arguments:
433
434 sub My::Object::FREEZE {
435 my ($self, $serialiser) = @_;
436
437 ($self->{type}, $self->{id}, $self->{variant})
438 }
439
440 sub My::Object::THAW {
441 my ($class, $serialiser, $type, $id, $variant) = @_;
442
443 $class-<new (type => $type, id => $id, variant => $variant)
444 }
445
446
268=head2 MAGIC HEADER 447=head1 MAGIC HEADER
269 448
270There is no way to distinguish CBOR from other formats 449There is no way to distinguish CBOR from other formats
271programmatically. To make it easier to distinguish CBOR from other 450programmatically. To make it easier to distinguish CBOR from other
272formats, the CBOR specification has a special "magic string" that can be 451formats, the CBOR specification has a special "magic string" that can be
273prepended to any CBOR string without changing it's meaning. 452prepended to any CBOR string without changing it's meaning.
276prepend this string tot he CBOR data it generates, but it will ignroe it 455prepend this string tot he CBOR data it generates, but it will ignroe it
277if present, so users can prepend this string as a "file type" indicator as 456if present, so users can prepend this string as a "file type" indicator as
278required. 457required.
279 458
280 459
460=head1 THE CBOR::XS::Tagged CLASS
461
462CBOR has the concept of tagged values - any CBOR value can be tagged with
463a numeric 64 bit number, which are centrally administered.
464
465C<CBOR::XS> handles a few tags internally when en- or decoding. You can
466also create tags yourself by encoding C<CBOR::XS::Tagged> objects, and the
467decoder will create C<CBOR::XS::Tagged> objects itself when it hits an
468unknown tag.
469
470These objects are simply blessed array references - the first member of
471the array being the numerical tag, the second being the value.
472
473You can interact with C<CBOR::XS::Tagged> objects in the following ways:
474
475=over 4
476
477=item $tagged = CBOR::XS::tag $tag, $value
478
479This function(!) creates a new C<CBOR::XS::Tagged> object using the given
480C<$tag> (0..2**64-1) to tag the given C<$value> (which can be any Perl
481value that can be encoded in CBOR, including serialisable Perl objects and
482C<CBOR::XS::Tagged> objects).
483
484=item $tagged->[0]
485
486=item $tagged->[0] = $new_tag
487
488=item $tag = $tagged->tag
489
490=item $new_tag = $tagged->tag ($new_tag)
491
492Access/mutate the tag.
493
494=item $tagged->[1]
495
496=item $tagged->[1] = $new_value
497
498=item $value = $tagged->value
499
500=item $new_value = $tagged->value ($new_value)
501
502Access/mutate the tagged value.
503
504=back
505
506=cut
507
508sub tag($$) {
509 bless [@_], CBOR::XS::Tagged::;
510}
511
512sub CBOR::XS::Tagged::tag {
513 $_[0][0] = $_[1] if $#_;
514 $_[0][0]
515}
516
517sub CBOR::XS::Tagged::value {
518 $_[0][1] = $_[1] if $#_;
519 $_[0][1]
520}
521
522=head2 EXAMPLES
523
524Here are some examples of C<CBOR::XS::Tagged> uses to tag objects.
525
526You can look up CBOR tag value and emanings in the IANA registry at
527L<http://www.iana.org/assignments/cbor-tags/cbor-tags.xhtml>.
528
529Prepend a magic header (C<$CBOR::XS::MAGIC>):
530
531 my $cbor = encode_cbor CBOR::XS::tag 55799, $value;
532 # same as:
533 my $cbor = $CBOR::XS::MAGIC . encode_cbor $value;
534
535Serialise some URIs and a regex in an array:
536
537 my $cbor = encode_cbor [
538 (CBOR::XS::tag 32, "http://www.nethype.de/"),
539 (CBOR::XS::tag 32, "http://software.schmorp.de/"),
540 (CBOR::XS::tag 35, "^[Pp][Ee][Rr][lL]\$"),
541 ];
542
543Wrap CBOR data in CBOR:
544
545 my $cbor_cbor = encode_cbor
546 CBOR::XS::tag 24,
547 encode_cbor [1, 2, 3];
548
281=head2 CBOR and JSON 549=head1 CBOR and JSON
282 550
283TODO 551CBOR is supposed to implement a superset of the JSON data model, and is,
552with some coercion, able to represent all JSON texts (something that other
553"binary JSON" formats such as BSON generally do not support).
554
555CBOR implements some extra hints and support for JSON interoperability,
556and the spec offers further guidance for conversion between CBOR and
557JSON. None of this is currently implemented in CBOR, and the guidelines
558in the spec do not result in correct round-tripping of data. If JSON
559interoperability is improved in the future, then the goal will be to
560ensure that decoded JSON data will round-trip encoding and decoding to
561CBOR intact.
284 562
285 563
286=head1 SECURITY CONSIDERATIONS 564=head1 SECURITY CONSIDERATIONS
287 565
288When you are using CBOR in a protocol, talking to untrusted potentially 566When you are using CBOR in a protocol, talking to untrusted potentially
356Please refrain from using rt.cpan.org or any other bug reporting 634Please refrain from using rt.cpan.org or any other bug reporting
357service. I put the contact address into my modules for a reason. 635service. I put the contact address into my modules for a reason.
358 636
359=cut 637=cut
360 638
361our $true = do { bless \(my $dummy = 1), "CBOR::XS::Boolean" };
362our $false = do { bless \(my $dummy = 0), "CBOR::XS::Boolean" };
363
364sub true() { $true }
365sub false() { $false }
366
367sub is_bool($) {
368 UNIVERSAL::isa $_[0], "CBOR::XS::Boolean"
369# or UNIVERSAL::isa $_[0], "CBOR::Literal"
370}
371
372XSLoader::load "CBOR::XS", $VERSION; 639XSLoader::load "CBOR::XS", $VERSION;
373
374package CBOR::XS::Boolean;
375
376use overload
377 "0+" => sub { ${$_[0]} },
378 "++" => sub { $_[0] = ${$_[0]} + 1 },
379 "--" => sub { $_[0] = ${$_[0]} - 1 },
380 fallback => 1;
381
3821;
383 640
384=head1 SEE ALSO 641=head1 SEE ALSO
385 642
386The L<JSON> and L<JSON::XS> modules that do similar, but human-readable, 643The L<JSON> and L<JSON::XS> modules that do similar, but human-readable,
387serialisation. 644serialisation.
388 645
646The L<Types::Serialiser> module provides the data model for true, false
647and error values.
648
389=head1 AUTHOR 649=head1 AUTHOR
390 650
391 Marc Lehmann <schmorp@schmorp.de> 651 Marc Lehmann <schmorp@schmorp.de>
392 http://home.schmorp.de/ 652 http://home.schmorp.de/
393 653
394=cut 654=cut
395 655
6561
657

Diff Legend

Removed lines
+ Added lines
< Changed lines
> Changed lines